Do I need a lab to practise?

No, the Paper 3 graph and analysis skills can be practised from home with example data.

What must a complete experiment plan include?

An aim, the manipulated, responding and constant variables, the apparatus with a labelled diagram, a step-by-step method that controls the constants and repeats readings, a planned data table, and how the results will be analysed.

Do I need a hypothesis?

Where the question asks for one, state a testable prediction linking the two variables, for example that a longer pendulum gives a longer period. It should be something your method can actually confirm or reject.

How do I make the test fair?

Change only the manipulated variable and hold every other factor constant, listing each one and how it is controlled. Repeating each reading and averaging reduces random error and strengthens the plan.

Science skill: Planning a complete experiment

Exam tip

This is a Paper 3 (practical) skill. Practise it from home with example data, no lab needed.

Aim

To learn how to plan and write up a complete experiment, from the aim and variables through to the method of analysis.

Variables

Manipulated: The plan names the one manipulated variable that will be changed.
Responding: The plan names the responding variable that will be measured.
Constant: The plan lists the constant variables and how they will be controlled.

Apparatus & materials

The aim or question to be investigated
Pen and paper for the plan and a labelled diagram
A list of the apparatus to be used

Procedure

State the aim, and where suitable an inference and a testable hypothesis.
List the variables: the manipulated variable, the responding variable, and the constant variables.
List the apparatus and materials, and describe the arrangement with a labelled diagram.
Describe the procedure step by step, saying how the manipulated variable is changed, how the responding variable is measured, and how readings are repeated.
Explain how the data will be tabulated, including any derived quantities.
Explain how the data will be analysed, what will be plotted, and how the gradient or shape leads to the conclusion.

Tabulating results

Plan the layout of the data table before collecting any readings, with the manipulated variable first, then the responding variable and any derived columns, each headed with its unit.

The graph

State clearly what will be plotted against what, and say what a straight line, its gradient, or its intercept would tell you about the relationship being investigated.

Analysis

Examiners award credit for a testable hypothesis, a complete set of variables, a workable procedure that controls the constant variables and repeats readings, and an analysis method that actually leads to the aim.

Precautions

Control every constant variable so the test is fair.
Include repeated readings so an average can be taken.
Make sure the method actually tests the hypothesis and leads to the aim.

Sample results and what they show

Suppose the aim is to plan an experiment on how the period of a pendulum depends on its length. A good plan predicts the shape of the results before any are taken.

An anticipated set of example readings (not real data) might be: length 0.20 m giving period 0.90 s, 0.40 m giving 1.27 s, 0.60 m giving 1.55 s, 0.80 m giving 1.80 s, and 1.00 m giving 2.01 s.

The period rises as the length rises, but not in equal steps, so the raw graph would curve. Planning for that, you decide in advance to calculate T² for each length so the analysis graph becomes a straight line.

The manipulated variable is length l in m.
The responding variable is period T in s, timed over many swings.
The derived column T² in s² is planned into the table from the start.

Thinking through the expected numbers shows the plan will actually work before you touch any apparatus.

Reading the graph and finding the answer

The plan should state exactly what will be plotted. Here you plot T² (y-axis) against l (x-axis), because squaring the period turns the curved raw relationship into a straight line through the origin.

The plan then says how the answer comes from the gradient. Using two points that would lie on the line, for example (0.20 m, 0.81 s²) and (1.00 m, 4.04 s²):

gradient = (4.04 s² − 0.81 s²) ÷ (1.00 m − 0.20 m) = 3.23 s² ÷ 0.80 m = 4.04 s² m⁻¹

Since T² = (4π²/g) l, the gradient equals 4π²/g, so g = 4π² ÷ gradient = 39.5 ÷ 4.04 s² m⁻¹ = 9.8 m s⁻². Writing this analysis step into the plan proves the chosen graph really leads back to the aim, which is what a complete plan must show.

Marks examiners look for

A planning question in Paper 3 is marked on the completeness of the plan, not on any actual readings.

State the aim and, where suitable, a testable hypothesis such as a longer pendulum has a longer period.
Name all three kinds of variable: the manipulated length, the responding period, and the constants such as the mass of the bob and the swing angle.
List the apparatus and give a labelled diagram of the arrangement.
Write a workable step-by-step method that controls the constants and repeats each timing to average out random error.

The science-process skills tested are making a hypothesis, controlling variables, and planning a fair, repeatable procedure. Finish by explaining how the data will be tabulated, including the derived T² column, and how the graph and its gradient lead to the value of g.

A plan that names the variables, keeps the test fair, and shows the analysis reaching the aim gains full credit.

Source: DSKP KSSM Physics Form 4 and 5 (Versi English) (Bahagian Pembangunan Kurikulum (BPK), KPM)

Written by the spmphysics.com.my editorial team.· Updated 5 Sept 2026